Book Success Prediction with Pretrained Sentence Embeddings and Readability Scores

TL;DR

This paper introduces a novel approach for predicting book success using pretrained sentence embeddings and readability scores, outperforming previous methods without relying on lexical or syntactic features.

Contribution

The study presents a new model that combines pretrained sentence embeddings with readability scores for book success prediction, requiring only the first 1,000 sentences.

Findings

Model outperforms strong baselines by up to 6.4% F1-score

Only the first 1,000 sentences are needed for accurate prediction

No count-based, lexical, or syntactic features are required

Abstract

Predicting the potential success of a book in advance is vital in many applications. This could help both publishers and readers in their decision-making process whether or not a book is worth publishing and reading, respectively. In this paper, we propose a model that leverages pretrained sentence embeddings along with various readability scores for book success prediction. Unlike previous methods, the proposed method requires no count-based, lexical, or syntactic features. Instead, we use a convolutional neural network over pretrained sentence embeddings and leverage different readability scores through a simple concatenation operation. Our proposed model outperforms strong baselines for this task by as large as 6.4\% F1-score points. Moreover, our experiments show that according to our model, only the first 1K sentences are good enough to predict the potential success of books.